350 



Profs. Liveing and Dewar 



[May 2, 



scopical examination directly, without any transference of the infusion ; 

 and the bulb E being filled with calcined air, enables the experimenter, 

 when the whole piece of apparatus is closed at J, to restore air to the 

 bulb A containing the infusion, by breaking the thin septum F by means 

 of the comparatively heavy piece of platinum H, which opens a commu- 

 nication between E and A. 

 Figure 27. Is the same as the above, except for the added tube MIS". This is 

 intended to be used if the fluid should prove to be sterilized, for the 

 purpose of inoculating it again by means of a piece of platinum which 

 is charged with the fluid in which the organism is abounding. This is 

 introduced into the tube, which is at once closed at M. It is then made 

 to break the septum L, and falls into the fluid inoculating it, to discover 

 if it still has the power to sustain the organism. 



All the drawings of the organism are magnified 3,000 diameters, except 

 where otherwise indicated on the plates. 



II. " On the Reversal of the Lines of Metallic Vapours." By G. 

 D. Liveing, M.A., Professor of Chemistry, and J. Dewar, 

 M.A., F.R.S., Jacksonian Professor, University of Cambridge. 

 No. II. Received March 26, 1878. 



Since our last communication to the Society we have succeeded in 

 reversing characteristic lines of the vapours of rubidium and ceesium. 

 Considering the known volatility of these elements, and the small 

 quantity of their compounds at our disposal, we thought it better to 

 try the effects first in glass tubes. For this purpose a piece of com- 

 bustion tubing had one end drawn out and the end turned up sharply, 

 and sealed off (like an ill-made combustion tube of the usual form) so 

 as to produce an approximately plane face at the end of the tube; a 

 small bulb was then blown at about an inch from the end, and the 

 tube drawn out at about an inch from the bulb on the other side, so as 

 to form a long narrower tube. Some dry rubidium or caesium chlo- 

 ride was next introduced into the bulb, and a fragment of fresh cut 

 sodium, and the narrow part of the tube turned up, so as to allow the 

 tube and bulb to be seen through in the direction of the axis of the tube. 

 The open end was then attached to a Sprengel pump, and the air ex- 

 hausted ; the sodium was then melted, and afterwards either dry 

 hydrogen or dry nitrogen admitted, and the end of the tube sealed off 

 at nearly the atmospheric pressure. We found it necessary to have 

 this pressure of gas inside the tube, otherwise the metal distilled so 

 fast on heating it that the ends were speedily obscured by condensed 

 drops of metal. Through these tubes placed lengthways in front of a 

 spectroscope, a lime light was viewed. On warming the bulb of a 

 tube in which rubidium chloride had been sealed up with sodium, the 

 D lines were of course very soon seen, and very soon there appeared two 

 dark lines near the extremity of the violet light, which, on measure- 



